Failure observations and mechanical modeling
Yuan Yuan, Min Ding, Sheila Chopin
Abstract
Yuan Yuan, Min Ding, Sheila Chopin
Abstract
Mechanical simulation and modeling play increasingly important roles in semiconductor industry. It provides guidelines on material selection, package geometry configuration, and package platform selection. It also provides insights and predictions on package reliability and failure mechanisms. Many mechanical models on reliability and failure issues aim directly at observed failures, such as solder cracks or die cracks. Simulations based on straightforward observations from failures are often correct, but not always. For example, failure analysis (FA) photos are usually taken at room temperature showing a static failure moment. This may not provide true “failure mode” at the failure temperature, and may miss the dynamic motion of the procedure. Therefore, the observed failures need to be carefully analyzed to discover the true failure mode and failure mechanism. These analyses may lead to different failure mechanism and result in simulations of different focus for improvements. This paper provides a few examples on semiconductor package failures and the analysis processes: how the FA photos lead to the discovery of the true failure mechanisms and how FEM simulation helped solving these issues.
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Mechanical simulation and modeling play increasingly important roles in semiconductor industry. It provides guidelines on material selection, package geometry configuration, and package platform selection. It also provides insights and predictions on package reliability and failure mechanisms. Many mechanical models on reliability and failure issues aim directly at observed failures, such as solder cracks or die cracks. Simulations based on straightforward observations from failures are often correct, but not always. For example, failure analysis (FA) photos are usually taken at room temperature showing a static failure moment. This may not provide true “failure mode” at the failure temperature, and may miss the dynamic motion of the procedure. Therefore, the observed failures need to be carefully analyzed to discover the true failure mode and failure mechanism. These analyses may lead to different failure mechanism and result in simulations of different focus for improvements. This paper provides a few examples on semiconductor package failures and the analysis processes: how the FA photos lead to the discovery of the true failure mechanisms and how FEM simulation helped solving these issues.
Key concepts: Computer science